CN115311443A - Oil leakage identification method for hydraulic pump - Google Patents
Oil leakage identification method for hydraulic pump Download PDFInfo
- Publication number
- CN115311443A CN115311443A CN202211245062.XA CN202211245062A CN115311443A CN 115311443 A CN115311443 A CN 115311443A CN 202211245062 A CN202211245062 A CN 202211245062A CN 115311443 A CN115311443 A CN 115311443A
- Authority
- CN
- China
- Prior art keywords
- oil leakage
- hydraulic pump
- area
- moment
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000002159 abnormal effect Effects 0.000 claims abstract description 88
- 238000003709 image segmentation Methods 0.000 claims abstract description 7
- 230000008859 change Effects 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 abstract description 2
- 239000003921 oil Substances 0.000 description 195
- 239000010720 hydraulic oil Substances 0.000 description 16
- 239000002184 metal Substances 0.000 description 7
- 238000001514 detection method Methods 0.000 description 5
- 230000002093 peripheral effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000009969 flowable effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
The invention relates to the field of visual identification methods, in particular to a hydraulic pump oil leakage identification method, which comprises the following steps: acquiring brightness difference images of each moment after the hydraulic pump is started; obtaining the oil leakage possibility of the hydraulic pump at each moment after starting according to the abnormal pixel points in the lightness difference image at each moment after starting the hydraulic pump; obtaining the oil leakage time of the hydraulic pump according to the oil leakage possibility; carrying out image segmentation on the brightness difference images at the oil leakage time and later by using a watershed algorithm to obtain all sub-regions in each brightness difference image; determining sub-regions corresponding to each other in each brightness difference image by using the center point coordinates of each sub-region in the brightness difference images at the oil leakage moment and later; and judging whether the sub-region is an oil leakage region according to whether the area of each corresponding sub-region in the brightness difference images at the continuous time is increased. The method is used for oil leakage identification of the hydraulic pump, and identification accuracy can be improved.
Description
Technical Field
The invention relates to the field of visual identification methods, in particular to a hydraulic pump oil leakage identification method.
Background
Hydraulic pumps leaking oil is one of the common failures of hydraulic pumps. When the oil leakage phenomenon occurs in the hydraulic pump, the output pressure of the system is unstable, the power is unstable, and the power of the executed part is deficient or unstable, so that accurate control cannot be performed; the power of the whole hydraulic pump is reduced, and the oil temperature is rapidly increased. If the oil leakage degree is serious, the control valve can not control the electric hydraulic pump, and further, the damage of elements is formed. In addition, leakage of hydraulic oil may cause fire, resulting in loss of personnel or industry. In order to solve this problem, the oil leakage detection is performed on the hydraulic pump before the hydraulic pump leaves the factory.
Most of the existing hydraulic pump oil leakage detection is manual detection, and constructors need to stop immediately to check if oil leakage is found, and then send the oil leakage to relevant departments to maintain.
However, with the increase of the production rhythm of the hydraulic pump, the detection pressure of the hydraulic pump is heavier and heavier, meanwhile, the oil liquid presents a transparent characteristic on the metal surface, and only an oil leakage area presents more specular reflection when oil leakage is distinguished from oil leakage, so that obvious color difference is avoided. And thus a great deal of effort is required for the inspector to perform the oil leakage inspection. In addition, because the oil leakage phenomenon is a gradually changing slow process, and small leakage is difficult to find by naked eyes in time in the detection process of a large number of hydraulic pumps, the invention designs a method for identifying the oil leakage phenomenon of the hydraulic pumps by using electronic equipment, and the method can effectively improve the identification efficiency and accuracy.
Disclosure of Invention
The invention provides a hydraulic pump oil leakage identification method, which comprises the following steps: acquiring brightness difference images of each moment after the hydraulic pump is started; obtaining the oil leakage possibility of the hydraulic pump at each moment after starting according to the abnormal pixel points in the lightness difference image at each moment after starting the hydraulic pump; obtaining the oil leakage time of the hydraulic pump according to the oil leakage possibility; carrying out image segmentation on the brightness difference images at the oil leakage time and later by using a watershed algorithm to obtain all sub-regions in each brightness difference image; determining sub-regions corresponding to each other in each brightness difference image by using the center point coordinates of each sub-region in the brightness difference images at the oil leakage moment and later; according to the method, whether the area of each corresponding subregion in the brightness difference image at continuous time is increased or not is judged, and compared with the prior art, the method is based on computer vision, a probability watershed algorithm is used, the probability that each pixel point in the image belongs to the oil leakage region is segmented according to the characteristic that the brightness of the metal surface of hydraulic oil can be increased, the oil leakage region is extracted, the oil leakage phenomenon of the hydraulic pump is identified, and the oil leakage position is determined according to the flowing direction of the hydraulic oil from the oil leakage point. The invention provides a method for identifying the oil leakage phenomenon of a hydraulic pump by using electronic equipment, which can effectively improve the identification efficiency and accuracy.
In order to achieve the purpose, the invention adopts the following technical scheme that the method for identifying the oil leakage of the hydraulic pump comprises the following steps:
and acquiring brightness difference images of the hydraulic pump to be detected at each moment after the hydraulic pump is started.
And taking the pixel points with the lightness difference value in the lightness difference image at each moment after the hydraulic pump is started as abnormal pixel points, and calculating the abnormal degree of each abnormal pixel point according to the lightness value of the abnormal pixel point and the average value of the lightness values of the pixel points in the eight neighborhoods of the abnormal pixel point.
Calculating the contribution amount of the abnormal pixel point to oil leakage judgment according to the distance from the abnormal pixel point to the oil pipe interface, the average value of the brightness values of the abnormal pixel point and the pixel points in eight neighborhoods of the abnormal pixel point; and calculating the oil leakage possibility of the hydraulic pump at each moment after starting based on the contribution amount of each abnormal pixel point.
And obtaining the oil leakage time of the hydraulic pump according to the oil leakage possibility at each time after the hydraulic pump is started.
And (3) carrying out image segmentation on the lightness difference images of the hydraulic pump at the oil leakage time and later by using a watershed algorithm to obtain all sub-areas in each lightness difference image.
And determining the sub-areas corresponding to each other in each lightness difference image by using the center point coordinates of each sub-area in the lightness difference images at the oil leakage time of the hydraulic pump and later.
And judging whether the sub-region is an oil leakage region according to whether the area of each corresponding sub-region in the brightness difference images at the continuous time is increased.
Further, in the method for identifying oil leakage of the hydraulic pump, the lightness difference image at each moment after the hydraulic pump to be detected is started is obtained as follows:
and collecting images at the interface of the oil pipe before and after the hydraulic pump to be detected is started.
And performing semantic segmentation on the image at the oil pipe interface to obtain the image of the area of the oil pipe interface before and after the hydraulic pump to be detected is started.
And converting the image of the oil pipe interface area into an HSV space, and acquiring brightness images before and after the hydraulic pump to be detected is started.
And (3) performing difference on brightness values of corresponding positions in the brightness image before the hydraulic pump to be detected is started and the brightness image after the hydraulic pump to be detected is started, and acquiring a brightness difference image at each moment after the hydraulic pump to be detected is started.
Further, the method for identifying oil leakage of a hydraulic pump, which calculates the contribution of the abnormal pixel point to oil leakage judgment according to the distance from the abnormal pixel point to the oil pipe interface, the abnormal pixel point and the brightness value mean value of the pixel points in the eight neighborhoods thereof, includes:
the calculation formula of the contribution amount is as follows:
in the formula ,
is shown as
The contribution of the abnormal degree of each abnormal pixel point to oil leakage judgment,
is as follows
The brightness value of each abnormal pixel point is calculated,
to be at the first
The mean value of the brightness values of all the pixel points in the eight neighborhoods taking the abnormal pixel points as the centers,
is as follows
The distance from each abnormal pixel point to the oil pipe interface.
Further, according to the method for identifying oil leakage of the hydraulic pump, the oil leakage time of the hydraulic pump is obtained as follows:
setting a threshold value, and judging the oil leakage possibility at each moment after the hydraulic pump is started: when the oil leakage possibility at each moment after the hydraulic pump is started is smaller than a threshold value, oil leakage does not exist at the moment, and oil leakage area identification is not needed; when the oil leakage possibility at each moment after the hydraulic pump is started is larger than or equal to the threshold value, the oil leakage area identification is needed when the oil leakage exists at the moment.
Further, in the method for identifying oil leakage of a hydraulic pump, the process of judging whether the sub-area is an oil leakage area is specifically as follows:
and taking the average value of oil leakage judgment contribution amounts corresponding to all pixel points in each subarea in the lightness difference image of the oil leakage moment of the hydraulic pump as the first suspected probability that each corresponding subarea belongs to the oil leakage area.
And performing iterative correction on the first suspected probability of each corresponding sub-region belonging to the oil leakage region according to the area difference between the corresponding sub-regions in the lightness difference image at the continuous moment to obtain the final suspected probability of each corresponding sub-region belonging to the oil leakage region.
Setting a threshold value, and judging the final suspected probability that each corresponding subregion belongs to the oil leakage region: and when the final suspected probability that the corresponding sub-area belongs to the oil leakage area is greater than the threshold value, the corresponding sub-area is the oil leakage area, the oil leakage area of the hydraulic pump to be detected is obtained, and the oil leakage position is determined.
Further, in the method for identifying oil leakage of a hydraulic pump, the expression for iteratively correcting the first suspected probability that each corresponding sub-region belongs to the oil leakage region is as follows:
in the formula ,
indicates the modified second
The frame image corresponds to the suspected probability that the jth sub-region belongs to the oil leakage region,
indicates the first before correction
The frame image corresponds to the suspected probability that the jth sub-area belongs to the oil leakage area,
is shown as
Frame image and
the area difference of the corresponding j sub-area in the frame image.
Further, in the method for identifying oil leakage of a hydraulic pump, the oil leakage position is determined as follows:
and acquiring a position change sequence of the central point of each corresponding sub-area in the lightness difference image of the hydraulic pump to be detected at the oil leakage moment and later.
And selecting a change sequence containing the central point of the oil leakage region from the position change sequence to obtain the initial position of the central point of the oil leakage region.
And determining the initial position of the central point of the oil leakage area as the oil leakage position of the hydraulic pump to be detected.
Further, the method for identifying oil leakage of a hydraulic pump, which calculates the oil leakage possibility of the hydraulic pump at each moment after starting based on the contribution of each abnormal pixel point, includes:
the calculation formula of the oil leakage possibility is as follows:
wherein ,
the possibility of oil leakage at each moment after the hydraulic pump is started;
denotes the first
Contribution of the abnormal degree of each abnormal pixel point to oil leakage judgment;
the number of abnormal pixel points in the brightness difference image at each moment after the hydraulic pump is started;
is an exponential function with a natural constant as a base.
The invention has the beneficial effects that:
based on computer vision, the probability watershed algorithm is used, according to the characteristic that the brightness of the metal surface of the hydraulic oil can be increased, the probability that each pixel point in the image belongs to the oil leakage area is segmented, the oil leakage area is extracted, the oil leakage phenomenon of the hydraulic pump is identified, and the oil leakage position is determined according to the flowing direction of the hydraulic oil from the oil leakage point. The invention provides a method for identifying the oil leakage phenomenon of a hydraulic pump by using electronic equipment, which can effectively improve the identification efficiency and accuracy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a hydraulic pump oil leakage identification method according to embodiment 1 of the present invention;
fig. 2 is a schematic flow chart of a hydraulic pump oil leakage identification method according to embodiment 2 of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Example 1
The embodiment of the invention provides an oil leakage identification method for a hydraulic pump, which comprises the following steps of:
s101, obtaining a brightness difference image of each moment after the hydraulic pump to be detected is started.
And obtaining a brightness difference image by subtracting the brightness value of each pixel point in the brightness image corresponding to the current moment image and the reference image.
S102, taking the pixel points with brightness difference values in the brightness difference image at each moment after the hydraulic pump is started as abnormal pixel points, and calculating the abnormal degree of each abnormal pixel point according to the brightness value of the abnormal pixel point and the average value of the brightness values of the pixel points in eight neighborhoods of the abnormal pixel point.
The larger the abnormal degree value is, the larger the probability that the abnormal pixel point belongs to the oil leakage area is.
S103, calculating to obtain the oil leakage possibility of the hydraulic pump at each moment after the hydraulic pump is started by using the abnormal degree of each abnormal pixel point and the distance between each abnormal pixel point and the oil pipe interface.
Wherein, the closer to the oil pipe interface, the greater the possibility of oil leakage.
And S104, obtaining the oil leakage time of the hydraulic pump according to the oil leakage possibility at each time after the hydraulic pump is started.
Wherein, setting a threshold value to obtain the oil leakage time.
And S105, carrying out image segmentation on the lightness difference images of the hydraulic pump at the oil leakage time and the subsequent time by using a watershed algorithm to obtain all sub-areas in each lightness difference image.
The watershed algorithm performs image segmentation according to local minimum values, and in the segmentation process, the similarity between adjacent pixels is used as an important reference, so that pixel points which are close in spatial position and have close gray values are connected with one another to form a closed contour.
S106, determining the sub-areas corresponding to each other in the lightness difference images by using the coordinates of the center point of each sub-area in the lightness difference images at the oil leakage time of the hydraulic pump and the time after the oil leakage time of the hydraulic pump.
Wherein, according to
The center point of each sub-region in the frame is in the previous frame (the first frame)
Frame) image is located in the same sub-region.
And S107, judging whether the sub-region is an oil leakage region according to whether the area of each corresponding sub-region in the brightness difference images at the continuous time is increased.
And the area of the corresponding sub-region is increased, so that the suspected probability of the sub-region is increased.
The beneficial effect of this embodiment lies in:
based on computer vision, the probability watershed algorithm is used, according to the characteristic that the brightness of the metal surface of the hydraulic oil can be increased, the probability that each pixel point in the image belongs to the oil leakage area is segmented, the oil leakage area is extracted, the oil leakage phenomenon of the hydraulic pump is identified, and the oil leakage position is determined according to the flowing direction of the hydraulic oil from the oil leakage point. The embodiment provides a method for identifying the oil leakage phenomenon of a hydraulic pump by using electronic equipment, and the identification efficiency and accuracy can be effectively improved.
Example 2
The main purposes of this embodiment are: the possibility of the oil leakage phenomenon is judged by using computer vision according to the change condition of the brightness degree of the metal surface; and further analyzing the image with high possibility of oil leakage and determining the oil leakage position, thereby realizing the automatic identification of the oil leakage phenomenon of the hydraulic pump and the automatic positioning of the oil leakage point.
The embodiment of the invention provides an oil leakage identification method for a hydraulic pump, which comprises the following steps of:
s201, collecting a surface image of the hydraulic pump.
Oil leak position appears easily in the hydraulic pump, like oil pipe kneck, set up camera 1, camera 2, camera 3, the interval angle between every camera is 120, wherein camera 1 places directly over the interface, camera 2,3 places respectively in the interface left and right sides, make three camera all just to oil pipe kneck, gather the image, because there is the backlight zone in the hydraulic pump under natural illumination, the image that leads to gathering is darker, influence discernment effect, consequently, each camera in this embodiment combines LED light filling lamp to discern, gather oil pipe interface area image, handle the image, realize oil leak discernment and the position determination of oil leak point according to the characteristic information in the image.
In the embodiment, an oil leakage identification and oil leakage point positioning device is designed, wherein the required electronic identification equipment comprises a camera, an LED light source and an embedded control system, an RGB camera is used for collecting an image of an oil pipe interface area, the embedded system is used for reading the image, an oil leakage area in the image is identified, and the oil leakage point position is determined according to the change condition of each area.
S202, acquiring an image of a hydraulic pump interface area.
The specific process of the step is as follows:
1. because the interior of the hydraulic pump is relatively complex, the image acquired by the camera also comprises other components, and in order to acquire an interface region, the DNN semantic segmentation network is used for identifying a target region in the embodiment; the specific process is as follows:
1) The network structure is an Encoder-Decoder structure, and the used data set is an oil pipe interface image data set of each angle;
2) The pixels needing to be segmented are divided into two types, namely the labeling process of the corresponding labels of the training set is as follows: the semantic label of the single channel, the pixel of the corresponding position belongs to the background class and is marked as 0, and the pixel of the corresponding position belongs to the interface area and is marked as 1.
3) The task of the network is classification, and all used loss functions are cross entropy loss functions.
2. After the hydraulic pump is placed, the cameras collect images for the first time, interface area recognition is carried out on the images collected by all the cameras by using a DNN network respectively, mask images of respective interface areas are obtained, connected area analysis is carried out on the obtained mask images respectively, the obtained connected areas are the interface areas collected by the cameras at all angles, and the interface area images collected for the first time are used as reference images for subsequent operation;
3. after the hydraulic pump is started, the camera acquires real-time images of the interface area and analyzes the subsequently acquired images.
4. Establishing a rectangular coordinate system by taking the image central point as a coordinate origin;
thus, the range of the hydraulic pump interface area is obtained.
And S203, preliminarily judging the possibility of oil leakage.
Since oil leakage does not occur at each oil leakage prone position on all the hydraulic pumps, in order to avoid unnecessary operations, the embodiment first preliminarily determines whether an oil leakage phenomenon exists in an image;
since the area where the hydraulic oil exists becomes specular reflection, the degree of brightness of the area with respect to the normal case increases, and therefore, when the degree of brightness in the image changes, there is a high possibility that an oil leak exists in the image. In consideration of metal reflection of an oil pipe interface, and the fact that the arranged light source is a point light source, the reflection degree of each position of an oil leakage area is different, so that although brightness changes in the position opposite to the camera, the change degree is small, and therefore threshold segmentation cannot be used for directly extracting the oil leakage area, namely the oil leakage area cannot be directly determined according to the change degree of the brightness.
The specific process is as follows:
1. because the oil leakage of the hydraulic pump can cause the brightness change, the reference image and the RGB image acquired at the current moment are respectively converted into HSV space, and the corresponding V-channel image, namely the brightness image, is acquired;
2. the brightness values of all pixel points in the brightness image corresponding to the current moment image and the reference image are subjected to difference to obtain a brightness difference image, and the pixel points with the brightness difference value in the image are abnormal pixel points;
3. preliminarily judging the possibility of oil leakage in the image according to the brightness values of the abnormal pixel points and the brightness mean values of the peripheral pixel points, wherein the specific process is as follows:
1) Due to the fact that noise exists in the image, the difference between the brightness value of the individual abnormal pixel and the brightness value of the peripheral pixels is large, and the oil leakage area is an area where the abnormal pixels are distributed in a concentrated mode, the abnormal degree of each abnormal pixel is obtained through calculation according to the brightness value of each abnormal pixel and the brightness average value of the pixels in eight neighborhoods of each abnormal pixel, and the larger the abnormal degree value is, the larger the probability that the abnormal pixel belongs to the oil leakage area is;
2) Since the location where the oil leakage occurs is the location of the joint between the oil pipe and the hydraulic pump, and the closer to the location, the greater the possibility of the oil leakage occurring, the closer to the joint location, the greater the reference degree for the oil leakage evaluation, the greater the distance is also required to be combined in the evaluation of the possibility of the oil leakage occurring in the whole image. Further, according to the abnormal pixel point, the oil is sent toCalculating contribution of the abnormal pixel points to oil leakage judgment by the distance of the pipe interface, the abnormal pixel points and the brightness value mean values of the pixel points in the eight neighborhoods of the abnormal pixel points; and calculating the oil leakage possibility of the hydraulic pump at each moment after starting based on the contribution amount of each abnormal pixel point. Thus the possibility of oil leakage
Can be expressed as:
wherein 
Characterization of
The abnormal degree of each abnormal pixel point;
is shown as
Contribution of the abnormal degree of each abnormal pixel point to oil leakage judgment;
the number of abnormal pixel points in the image is obtained;
is as follows
The distance from each abnormal pixel point to the original point is used for expressing the reference weight of the abnormal pixel point;
is as follows
The brightness value of each abnormal pixel point;
to be under the first
The lightness mean value of each pixel point in eight neighborhoods taking each abnormal pixel point as the center;
is an exponential function with a natural constant as a base.
3) When the temperature is higher than the set temperature
In the time, the possibility of oil leakage in the image is considered to be high, the oil leakage area needs to be identified, the position of an oil leakage point is determined, and the initial acquisition time of the current camera is recorded.
And S204, acquiring an oil leakage area.
In the embodiment, a watershed algorithm is used for segmenting abnormal pixel points in an image, the watershed algorithm is used for segmenting the image according to a local minimum value, and in the segmentation process, the similarity between adjacent pixels is used as an important reference, so that the pixel points which are close in spatial position and have close gray values are connected with each other to form a closed contour. The basic idea is to regard the image as a topological geomorphology, the gray value of each pixel of a point in the image represents the altitude of the point, each local minimum value and the influence area thereof are called a catchment basin, and the boundary of the catchment basin forms a watershed.
However, considering that the traditional watershed algorithm is sensitive to noise in an image and slight change of data and is easy to generate an over-segmentation phenomenon, the obtained abnormal degree of each pixel point not only combines the abnormal brightness value of the pixel point, but also combines the brightness values of the peripheral pixel points, so that the abnormal degree of the abnormal pixel point can be considered as that the brightness value of the current pixel point is corrected by combining the brightness values of the peripheral pixel points, the watershed segmentation is performed on the abnormal degree value of each pixel point, the noise point can be eliminated, and the over-segmentation phenomenon is avoided.
Since the hydraulic oil is flowable and is subject to extension during leakage, this embodiment is based on the changing characteristics of the hydraulic oil versus the continuity
Analyzing the frame image, wherein the specific process is as follows:
1. to the first
The analysis process of the frame image is as follows:
1) Performing image segmentation on the obtained brightness difference image by using a watershed algorithm to obtain a plurality of sub-regions, wherein each sub-region corresponds to a similar brightness region;
2) Because the brightness values of the pixel points in the oil-free region have certain similarity, the obtained sub-regions contain oil-free normal regions, and the probability that each sub-region belongs to the oil-leakage region needs to be calculated, then
First of frame image
Suspected probability of a sub-region belonging to an oil-leaking region
Can be expressed as
The average value of oil leakage judgment contribution amounts corresponding to all pixel points in each sub-area is subjected to normalization to obtain a result;
3) For each sub-regionA smallest surrounding frame, wherein the intersection point of the diagonal lines of the surrounding frame is taken as the center point coordinate of each sub-region, the area of each sub-region and the suspected probability of belonging to the oil leakage region are combined to obtain a triple describing the characteristics of each sub-region, and then the third step is to calculate the third step
The feature triple corresponding to each sub-region can be expressed as
, wherein 
Is a first
In the frame image
The coordinates of the center point of each sub-region,
are respectively the first
The area of the sub-region and the suspected probability of belonging to the oil leakage region;
2. due to the fluidity of the hydraulic oil, the hydraulic oil can extend downwards under the influence of gravity, so that the area of an oil leakage area is increased, and the area of a normal area is reduced along with the flowing of the hydraulic oil, so that the method is used for continuously obtaining the oil
The brightness difference image corresponding to the frame image is subjected to an abnormal brightness subregion, and the suspected probability that the region belongs to the oil leakage region is corrected according to the area change of each subregion of the adjacent frame image, and the specific process is as follows:
1) According to the first
The center point of each sub-region in the frame is in the previous frame (the first frame)
Frame) image is located in the same subarea;
2) Calculating the area difference between every two corresponding subregions in the images of the adjacent frames, and correcting according to the change condition of the area difference, wherein the correction process is as follows:
wherein 
Represents the current frame (the first
Frame) and the previous frame (the first frame)
Frame) the area difference of the corresponding sub-region in the image;
when in use
When the value is more than 0, the area of the sub-region is increased, and the correspondence is realized
Is (0,1) and follows
Such that the probability of suspicion of the sub-region increases; when in use
When the value is less than 0, the area of the sub-region is increased, and the correspondence is realized
Is in the range of (-1,0) and follows
Is increased, so that the probability of suspicion of the sub-region is decreased.
3) When in use
When is shown as
In the frame of
The area of the sub-region is changed, and the center point of the sub-region is updated to be the current first point
The coordinates of the center point of the sub-region in the frame.
4) Sequentially processing each subarea until all the subareas are processed;
3. repeating the above operation on the obtained continuous frame images to obtain the final suspected probability of each subregion, judging the oil leakage region according to the suspected probability of each subregion in the last frame image, and setting a threshold value
And when the suspected probability of each sub-area is greater than the threshold value, the area is considered as an oil leakage area.
Thus, an oil leakage area is obtained.
And S205, further judging the oil leakage position according to the distribution situation of the oil leakage area.
In order to determine whether an oil leakage is occurring at the same oil leakage location, a further determination is made of the oil leakage location in the oil leakage area.
The oil leakage position of the oil leakage area extracted in the step is determined according to the distribution condition of the hydraulic oil expansion area and the gray level characteristics of each expansion area; the oil leakage phenomenon in the images collected by the cameras on the two sides is probably caused by that the oil leakage point is positioned above the camera collection area and flows downwards, so that the three cameras need to be judged respectively.
The specific process is as follows:
1. acquiring a position change sequence of the central point of each sub-region in the continuous n frames of images, selecting the change sequence containing the central point of the final oil leakage region, and determining the initial oil leakage position in the corresponding camera acquisition range;
2. within the image range collected by the camera 1, the initial oil leakage position is an oil leakage point;
3. for the image captured by camera 2,3:
when the initial sub-region corresponding to each initial oil leakage position does not contain the pixel point with the maximum vertical coordinate of the whole image (namely does not contain the upper edge of the image), each initial oil leakage position corresponds to one oil leakage point;
recording initial acquisition time of the three cameras in an initial sub-area corresponding to the initial oil leakage position as initial acquisition time
When the camera 2 is satisfied
The camera 3 satisfies
Each initial oil leakage position corresponds to an oil leakage point; otherwise, the initial oil leakage level is regarded as a non-oil leakage point, and the oil leakage point is removed.
The beneficial effect of this embodiment lies in:
based on computer vision, the probability watershed algorithm is used, according to the characteristic that the brightness of the metal surface of the hydraulic oil can be increased, the probability that each pixel point in the image belongs to the oil leakage area is segmented, the oil leakage area is extracted, the oil leakage phenomenon of the hydraulic pump is identified, and the oil leakage position is determined according to the flowing direction of the hydraulic oil from the oil leakage point. The embodiment provides a method for identifying the oil leakage phenomenon of a hydraulic pump by using electronic equipment, and the identification efficiency and accuracy can be effectively improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. A hydraulic pump oil leakage identification method is characterized by comprising the following steps:
acquiring brightness difference images of the hydraulic pump to be detected at each moment after the hydraulic pump to be detected is started;
taking the pixels with lightness difference values in the lightness difference image at each moment after the hydraulic pump is started as abnormal pixels, and calculating the abnormal degree of each abnormal pixel according to the lightness value of the abnormal pixel and the average value of the lightness values of the pixels in eight neighborhoods of the abnormal pixel;
calculating the contribution of the abnormal pixel points to oil leakage judgment according to the distance from the abnormal pixel points to the oil pipe interface, the average value of the brightness values of the abnormal pixel points and the pixel points in the eight neighborhoods of the abnormal pixel points; calculating the oil leakage possibility of each moment after the hydraulic pump is started based on the contribution amount of each abnormal pixel point;
obtaining the oil leakage time of the hydraulic pump according to the oil leakage possibility of the hydraulic pump at each time after the hydraulic pump is started;
performing image segmentation on lightness difference images of the hydraulic pump at the oil leakage time and later by using a watershed algorithm to obtain all sub-areas in each lightness difference image;
determining sub-areas corresponding to each other in each lightness difference image by using the center point coordinates of each sub-area in the lightness difference image at the oil leakage moment of the hydraulic pump and at the moment after the oil leakage moment of the hydraulic pump;
and judging whether the sub-region is an oil leakage region according to whether the area of each corresponding sub-region in the brightness difference images at the continuous time is increased.
2. The hydraulic pump oil leakage identification method according to claim 1, wherein the lightness difference image at each moment after the hydraulic pump to be detected is started is obtained as follows:
collecting images at the interface of an oil pipe before and after the hydraulic pump to be detected is started;
performing semantic segmentation on an image at an oil pipe interface to obtain an image of the area of the oil pipe interface before and after the hydraulic pump to be detected is started;
converting the image of the oil pipe interface area into an HSV space, and acquiring brightness images before and after starting of a hydraulic pump to be detected;
and (3) performing difference on brightness values of corresponding positions in the brightness image before the hydraulic pump to be detected is started and the brightness image after the hydraulic pump to be detected is started, and acquiring a brightness difference image at each moment after the hydraulic pump to be detected is started.
3. The method for identifying oil leakage of a hydraulic pump according to claim 1, wherein the step of calculating the contribution of the abnormal pixel point to oil leakage judgment according to the distance from the abnormal pixel point to the oil pipe interface, the average value of the brightness values of the abnormal pixel point and the pixel points in eight neighborhoods of the abnormal pixel point comprises the following steps:
the calculation formula of the contribution amount is as follows:
in the formula ,
is shown as
The contribution of the abnormal degree of each abnormal pixel point to oil leakage judgment is
The brightness value of each abnormal pixel point is calculated,
to be under the first
The mean value of the brightness values of all the pixel points in the eight neighborhoods taking the abnormal pixel points as the centers,
is as follows
The distance from each abnormal pixel point to the oil pipe interface.
4. The method for identifying oil leakage of a hydraulic pump according to claim 1, wherein the oil leakage time of the hydraulic pump is obtained as follows:
setting a threshold value, and judging the oil leakage possibility at each moment after the hydraulic pump is started: when the oil leakage possibility at each moment after the hydraulic pump is started is smaller than a threshold value, oil leakage does not exist at the moment, and oil leakage area identification is not needed; when the oil leakage possibility at each moment after the hydraulic pump is started is larger than or equal to the threshold value, oil leakage area identification is needed when oil leakage exists at the moment.
5. The method for identifying oil leakage of a hydraulic pump according to claim 1, wherein the process of judging whether the sub-area is an oil leakage area is specifically as follows:
taking the average value of oil leakage judgment contribution amounts corresponding to all pixel points in each subarea in the lightness difference image of the oil leakage moment of the hydraulic pump as a first suspected probability that each corresponding subarea belongs to the oil leakage area;
performing iterative correction on the first suspected probability of each corresponding sub-region belonging to the oil leakage region according to the area difference between the corresponding sub-regions in the lightness difference image at the continuous moment to obtain the final suspected probability of each corresponding sub-region belonging to the oil leakage region;
setting a threshold value, and judging the final suspected probability that each corresponding subregion belongs to the oil leakage region: and when the final suspected probability that the corresponding sub-area belongs to the oil leakage area is greater than the threshold value, the corresponding sub-area is the oil leakage area, the oil leakage area of the hydraulic pump to be detected is obtained, and the oil leakage position is determined.
6. The hydraulic pump oil leakage identification method according to claim 5, wherein the expression for iteratively correcting the first suspected probability that each corresponding sub-area belongs to the oil leakage area is as follows:
in the formula ,
indicating modified the second
The frame image corresponds to the suspected probability that the jth sub-area belongs to the oil leakage area,
indicates the first before correction
The frame image corresponds to the suspected probability that the jth sub-area belongs to the oil leakage area,
denotes the first
Frame image and
the area difference of the corresponding j sub-area in the frame image.
7. The hydraulic pump oil leakage identification method according to claim 5, wherein the oil leakage position is determined as follows:
acquiring a position change sequence of the central point of each corresponding subregion in the lightness difference image of the hydraulic pump to be detected at the oil leakage moment and later;
selecting a change sequence containing the central point of the oil leakage region from the position change sequence to obtain an initial position of the central point of the oil leakage region;
and determining the initial position of the central point of the oil leakage area as the oil leakage position of the hydraulic pump to be detected.
8. The method for identifying oil leakage of a hydraulic pump according to claim 1, wherein the calculating of the oil leakage possibility at each moment after the hydraulic pump is started based on the contribution amount of each abnormal pixel point comprises:
the calculation formula of the oil leakage possibility is as follows:
wherein ,
the oil leakage possibility at each moment after the hydraulic pump is started; is shown as
Contribution of the abnormal degree of each abnormal pixel point to oil leakage judgment;
the number of abnormal pixel points in the brightness difference image at each moment after the hydraulic pump is started;
is an exponential function with a natural constant as a base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211245062.XA CN115311443B (en) | 2022-10-12 | 2022-10-12 | Oil leakage identification method for hydraulic pump |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211245062.XA CN115311443B (en) | 2022-10-12 | 2022-10-12 | Oil leakage identification method for hydraulic pump |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115311443A true CN115311443A (en) | 2022-11-08 |
| CN115311443B CN115311443B (en) | 2023-05-23 |
Family
ID=83868028
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211245062.XA Active CN115311443B (en) | 2022-10-12 | 2022-10-12 | Oil leakage identification method for hydraulic pump |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115311443B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117635610B (en) * | 2024-01-25 | 2024-04-09 | 青岛正大正电力环保设备有限公司 | Visual detection method for oil leakage of oil pipe of hydraulic tensioning mechanism |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016090560A (en) * | 2014-10-29 | 2016-05-23 | 株式会社日立製作所 | Oil leak detector and oil leak detection method |
| CN113095114A (en) * | 2019-12-23 | 2021-07-09 | 新疆金风科技股份有限公司 | Method, device and equipment for detecting oil leakage of hydraulic system |
| CN114723691A (en) * | 2022-03-28 | 2022-07-08 | 江苏新之阳新能源科技有限公司 | Method for detecting oil leakage fault degree of hydraulic system based on artificial intelligence |
| CN115100451A (en) * | 2022-08-29 | 2022-09-23 | 启东万惠机械制造有限公司 | Data expansion method for monitoring oil leakage of hydraulic pump |
-
2022
- 2022-10-12 CN CN202211245062.XA patent/CN115311443B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016090560A (en) * | 2014-10-29 | 2016-05-23 | 株式会社日立製作所 | Oil leak detector and oil leak detection method |
| CN113095114A (en) * | 2019-12-23 | 2021-07-09 | 新疆金风科技股份有限公司 | Method, device and equipment for detecting oil leakage of hydraulic system |
| CN114723691A (en) * | 2022-03-28 | 2022-07-08 | 江苏新之阳新能源科技有限公司 | Method for detecting oil leakage fault degree of hydraulic system based on artificial intelligence |
| CN115100451A (en) * | 2022-08-29 | 2022-09-23 | 启东万惠机械制造有限公司 | Data expansion method for monitoring oil leakage of hydraulic pump |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117635610B (en) * | 2024-01-25 | 2024-04-09 | 青岛正大正电力环保设备有限公司 | Visual detection method for oil leakage of oil pipe of hydraulic tensioning mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115311443B (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109816678B (en) | Automatic nozzle atomization angle detection system and method based on vision | |
| CN114419025A (en) | Fiberboard quality evaluation method based on image processing | |
| CN110148130B (en) | Method and device for detecting part defects | |
| CN109829914B (en) | Method and device for detecting product defects | |
| CN105139386B (en) | A kind of image processing method of fast automatic detecting electric connector solder joint defective work | |
| CN115601367A (en) | LED lamp wick defect detection method | |
| CN112837290B (en) | Crack image automatic identification method based on seed filling algorithm | |
| CN115908411B (en) | Concrete curing quality analysis method based on visual detection | |
| CN113781402A (en) | Method and device for detecting chip surface scratch defects and computer equipment | |
| CN115082466B (en) | PCB surface welding spot defect detection method and system | |
| CN113592828B (en) | Nondestructive testing method and system based on industrial endoscope | |
| CN111814686A (en) | Vision-based power transmission line identification and foreign matter invasion online detection method | |
| CN111724375B (en) | Screen detection method and system | |
| CN115131354B (en) | Laboratory plastic film defect detection method based on optical means | |
| CN110175556B (en) | Remote sensing image cloud detection method based on Sobel operator | |
| CN116152242B (en) | Visual detection system of natural leather defect for basketball | |
| CN113324864A (en) | Pantograph carbon slide plate abrasion detection method based on deep learning target detection | |
| CN113487563B (en) | EL image-based self-adaptive detection method for hidden cracks of photovoltaic module | |
| CN112686264A (en) | Digital instrument reading method and device, computer equipment and storage medium | |
| CN111665199A (en) | Wire and cable color detection and identification method based on machine vision | |
| CN115311443A (en) | Oil leakage identification method for hydraulic pump | |
| CN116883408A (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN118097305A (en) | Method and system for detecting quality of semiconductor light-emitting element | |
| CN113888503A (en) | Product appearance detection method and device and storage medium | |
| CN117351472A (en) | Tobacco leaf information detection method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |